Distilling Directional Verification

arXiv:2610.00997 · cs.CL, cs.AI, cs.LG · Submitted 2026-10-01 · Read on arXiv

cs.CL, cs.AI, cs.LG

Submitted: 2026-10-01

Updated: 2026-10-01

Code: https://github.com/js-lee-AI/directional-verification

Terminology

• Accuracy — 138× in this paper · explained in 1 episode
• Distillation — 21× in this paper · explained in 2 episodes
• Aggregation — 16× in this paper · explained in 1 episode
• Verification — 12× in this paper · explained in 2 episodes
• Adaptation — 9× in this paper · explained in 1 episode
• In International Conference on Learning Representations — 7× in this paper
• Verifier — 6× in this paper · explained in 2 episodes
• Diffusion Model — 5× in this paper · explained in 8 episodes
• Diffusion Models — 5× in this paper · explained in 6 episodes
• Knowledge Distillation — 4× in this paper · explained in 11 episodes
• Empirical Methods in Natural Language Processing — 2× in this paper
• Fidelity — 2× in this paper · explained in 2 episodes
• In Conference on Language Modeling — 2× in this paper
• On-Policy Distillation — 2× in this paper · explained in 3 episodes
• Transformer — 2× in this paper · explained in 2 episodes
• Alignment — 1× in this paper · explained in 1 episode
• Calibration — 1× in this paper · explained in 7 episodes
• Clustering — 1× in this paper · explained in 2 episodes
• Decode — 1× in this paper · explained in 2 episodes
• Fine-Tuning — 1× in this paper · explained in 5 episodes
• Foundation Model — 1× in this paper · explained in 3 episodes
• Generalization — 1× in this paper · explained in 13 episodes
• Low-Rank Adaptation (LoRA) — 1× in this paper · explained in 9 episodes
• Pseudo-labels — 1× in this paper · explained in 2 episodes
• Regularization — 1× in this paper · explained in 3 episodes
• Supervised Fine-Tuning (SFT) — 1× in this paper · explained in 8 episodes

Sources

• On-Policy Distillation of Language Models: Learning from Self-Generated Mistakes
• Structured Denoising Diffusion Models in Discrete State-Spaces
• Training Verifiers to Solve Math Word Problems
• RAFT: Reward rAnked FineTuning for Generative Foundation Model Alignment
• The Llama 3 Herd of Models
• Reinforced Self-Training (ReST) for Language Modeling
• Mistral 7B
• Exploring Knowledge Purification in Multi-Teacher Knowledge Distillation for LLMs
• Sequence-Level Knowledge Distillation
• FUSE: Ensembling Verifiers with Zero Labeled Data
• DART: Draft-Agreement Routing for Training-Free Adaptive Thinking Budgets in Hybrid Reasoning Models
• Answer-Conditioned Chains of Thought Degrade Verifiable-Reasoning Distillation in Large Language Models
• Let's Verify Step by Step
• Breaking the Reversal Curse in Autoregressive Language Models via Identity Bridge
• Large Language Diffusion Models
• Cross-Tokenizer Likelihood Scoring Algorithms for Language Model Distillation
• Beyond Human Data: Scaling Self-Training for Problem-Solving with Language Models
• 2 OLMo 2 Furious
• Qwen3 Technical Report

Related papers

• Exploring Solution Divergence and Its Effect on Large Language Model Problem Solving
• Ishigaki-IDS-Bench: A Benchmark for Generating Information Delivery Specification from BIM Information Requirements
• Subliminal Steering: Stronger Encoding of Hidden Signals
• MedStruct-S: A Benchmark for Key Discovery, Key-Conditioned QA and Semi-Structured Extraction from OCR Clinical Reports
• The End of Transformers? On Challenging Attention and the Rise of Sub-Quadratic Architectures
• Untangling the Mechanisms of Misleading Context in Medical Question Answering